US 11402402

I'll search for this specific patent number and check for any litigation or CAFC activity.

Let me check specifically for litigation and CAFC activity, plus verify the USPTO record.

Let me run a couple more targeted checks on any litigation/PTAB activity involving this patent's owner and the exact number.

US 11,402,402 B2 — Verified Summary

Confidence note: Bibliographic data below is corroborated by the authoritative full patent text supplied with this task and by Google Patents and Justia listings retrieved via search. I was unable to find any district-court or CAFC docket activity naming this patent (details in the "Litigation / CAFC" section — this is absence of evidence, not proof of non-existence).

Bibliographic Data

Abstract (as issued)

"The present disclosure provides substantially magnetometer-free systems and methods for motion capture of a subject including 3-D localization and posture tracking by fusing inertial sensors with a localization system and a biomechanical model of the subject. Using the novel Kalman filter based fusion techniques disclosed herein, the localization data aided with the biomechanical model can eliminate the drift in inertial yaw angle estimation."

Plain-Language Overview of the Independent Claims

The patent has two independent claims — claim 1 (system) and claim 6 (method). Claims 2–5 depend from claim 1. (I observed claims 1–6 in the published claim set; I did not verify that no further claims exist beyond claim 6.)

Claim 1 — System for capturing lower body motion. A wearable rig plus a processor:

• Seven IMUs — one each on the left foot, right foot, left shank, right shank, left thigh, right thigh, and pelvis. Each IMU has an accelerometer and a gyroscope and outputs rate-of-turn and acceleration signals.
• Three localization tags — mounted at the left mid-foot, right mid-foot, and waist.
• A localization sensor system that detects each tag and outputs position signals.
• A processing system that receives those signals, derives velocity signals from the position signals, and runs three chained Kalman filters plus a reconstruction stage:
  • a tilt Kalman filter producing pitch and roll for all seven segments;
  • a localization Kalman filter that fuses rate-of-turn, acceleration, tag position/velocity, and the tilt filter's pitch/roll for the feet and pelvis, and outputs position vectors for the three tag points (left mid-foot, right mid-foot, waist) and rotation matrices for the feet and pelvis;
  • a yaw Kalman filter that takes the shank/thigh pitch and roll from the tilt filter plus the position vectors and rotation matrices from the localization filter, and outputs yaw angles for the shanks and thighs;
  • a motion reconstruction block that reconstructs full three-dimensional motion of the subject's lower body from those outputs.

Claim 6 — Method for capturing motion of a subject (the generalized/kinematic-chain formulation). Steps: mount IMUs on body segments that form one or more kinematic chains (each chain having a root segment and an end segment); mount one localization tag at a localization point on each root and end segment; detect the tags to generate position and velocity signals; run the acceleration/rate-of-turn signals through a tilt Kalman filter to get pitch and roll for each segment; run the root/end-segment inertial signals plus position and velocity signals plus the tilt-filter pitch/roll through a localization Kalman filter to get a position vector per localization point and a rotation matrix per root/end segment; run the remaining segments' pitch/roll plus the localization-filter outputs through a yaw Kalman filter to get yaw angles for the segments that are neither root nor end; and reconstruct full 3-D motion from all three filters' outputs.

Dependent Claims (for completeness)

• Claim 2 — the reconstruction block determines a corrected waist position from the mid-foot position vectors and the segment pitch/roll/yaw, then reconstructs the left leg and right leg as two kinematic chains rooted at the corrected waist position.
• Claim 3 — the corrected waist position is the average of a left-leg-derived waist position and a right-leg-derived waist position.
• Claim 4 — each IMU is a tri-axial accelerometer and tri-axial gyroscope.
• Claim 5 — each IMU is initialized during a pre-capture initialization phase, and magnetic data is used only during that initialization phase (the claim that hard-codes the "magnetometer-free during operation" concept).

Technical gist of the disclosure

The core asserted novelty, per the specification, is twofold: (1) the UWB (or equivalent absolute localization, e.g. vision, accuracy better than ~15 cm) position data is used not only for position tracking but also to aid yaw estimation in a loosely-coupled filter, and (2) inertial data for some lower-body segments is aided by foot/waist position data to give magnetometer-free 3-D posture tracking. The yaw filter exploits the fact that, with known thigh/shank inclination, the horizontal components of the hip-to-ankle vector depend on the leg's yaw orientation, letting yaw be corrected without a magnetometer. Reported results: 3-D localization better than ~4.5 cm; knee-angle tracking error ~3.5° (low dynamic) to ~4.5° (high dynamic); ~20 s convergence time for the arbitrary initial yaw. Appendix A derives Eqs. (28)–(29) used to build the yaw filter's measurement matrix.

Litigation / CAFC 2026 Dockets

I ran multiple targeted searches and found no district-court complaint, IPR/PGR, or Federal Circuit appeal involving US 11,402,402 or its assignee BigMotion Technologies Inc. My searches for "402"-numbered patent litigation returned only unrelated patents, which I am flagging so they are not conflated with this one:

• U.S. 11,562,402 and 10,594,823 — RavenWhite Licensing v. Walmart, E.D. Tex. 2:24-cv-00688 (browser-cookie/ad-bidding patents) — different patent.
• U.S. 11,597,402 — Granite Vehicle Ventures v. Tesla, E.D. Tex. (autonomous-driving-mode patents) and related IPR2025-00943 — different patent.
• U.S. 10,148,402 — ASUS Technology Licensing v. AT&T (E.D. Tex. 2:23-cv-486) — different patent.
• The Wi-LAN " '402 patent" (V-Chip) in the Hon Hai matter — a much older, different patent.

Caveats on the litigation finding: (a) free web search is not a substitute for a Docket Navigator / PACER / PTAB API query, and (b) a recently filed (late 2025 or 2026) case or a sealed matter could exist without surfacing. If you need a definitive litigation-clearance answer, I'd confirm directly against PACER and the PTAB Patent Trial Case Tracking System under both the patent number and "BigMotion Technologies Inc." I also cannot verify whether the patent remains in force or has been reassigned since the last recorded 2018 assignment, and whether maintenance fees are paid — treat the "Active / expires 2037-07-03" entry as the register's statement rather than a legal conclusion.

I'll search for litigation involving US patent 11402402 across patent litigation databases.

Let me search more targeted litigation databases.

Summary: No litigation found for US 11,402,402

Based on my searches of patent-litigation and patent-trial resources (Unified Patents portal, Docket Alarm/PTAB records, CAFC opinions/orders, CourtListener/ PACER-style dockets, and Google Patents' "litigation" and "related proceedings" sections), I could find no district court litigation and no PTAB (IPR/PGR/CBM) proceeding involving U.S. Patent No. 11,402,402 ("Systems and methods for human body motion capture," Bigmotion Technologies Inc.)

I am stating this as "none found," not as a verified "none exists" — see the caveats below.

What the record does show for this patent

Notably, the Google Patents record for '402 lists only the assignment to Bigmotion Technologies Inc. and no "Litigation" or "Related Proceedings" entries — a strong (though not conclusive) indicator that the patent has not been asserted.

Cases I checked and ruled out (do NOT confuse these with '402)

1. Merck v. Wyeth (pneumococcal vaccine '024 patent) — the complaint discusses a "'024 patent," but that is a different Wyeth patent on pneumococcal conjugate vaccines (declaratory-judgment counts re: non-infringement/invalidity). It is not US 11,402,402. Do not conflate.
2. Covves LLC v. BigMouth Inc., No. 2:18-cv-08518 (C.D. Cal.) — surfaced because "BigMouth Inc." resembles "Bigmotion," but the case involved design patents on unicorn pool floats. Irrelevant.
3. 11402402 as a product code — the number also appears as a "Malleable Hexagon Reducing Bush" part number in a pipe-fittings catalogue. Irrelevant.
4. Unified Patents / Ozmo, Ideahub, MasterObjects, etc. — surfaced in PTAB/Unified searches but concern unrelated patents (e.g., '991, '228, '073). Not '402.

Possible corporate-context lead (unverified)

One search hit (for a different application, US 20210311546A1) showed a chain-of-title entry assigning rights to Brave Virtual Worlds, LLC / Brave Virtual Worlds, Inc. (Virginia). I could not confirm any link between Brave Virtual Worlds and US 11,402,402; the '402 page itself still lists Bigmotion Technologies Inc. as current assignee. Treat this as an unverified lead for further investigation, not a finding.

Caveats and recommended verification

My external searching returned substantial noise, so absence of results should be treated cautiously. To definitively confirm status, check:

• USPTO Patent Center / Assignment database for '402 (recordation + any security interests or re-assignments).
• USPTO PTAB (IPR/PGR/CBM search by patent number) — I found no post-grant challenge.
• Unified Patents litigation portal — https://portal.unifiedpatents.com/litigation/caselist (search by patent number).
• PACER / CourtListener (RECAP) — search "11,402,402" and "Bigmotion Technologies" as party.
• CAFC opinions/orders — no appeal found referencing this patent.

Bottom line

No litigation is known for US 11,402,402. As of the available records, the patent appears to be a non-asserted, currently Active asset held by Bigmotion Technologies Inc. (adjusted expiration 2037-07-03). If you need a litigation-grade clearance opinion, the checks listed above should be run directly against Docket Navigator, PACER, and Unified Patents, since my search coverage here was incomplete and returned only tangential matches.

I'll verify the PTAB landscape with targeted searches before writing the analysis.

Proceedings overview

No AIA trial proceedings are on file for US11402402. The structured USPTO Open Data Portal block for this patent returns zero IPR, PGR, or CBM petitions, and my independent web checks of PTAB-facing sources (ptacts.uspto.gov, Google Patents "Similar/Related" litigation data, and general queries pairing the patent number and patent owner "Bigmotion Technologies Inc." with "IPR," "PTAB," and "petition") surfaced nothing beyond prosecution-history documents. Total count: 0 proceedings (0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution-denied). Bottom-line defensive posture for a defendant: this is not a hardened patent — it is an untested one. Every claim of US11402402 remains live, never construed by the Board, and the PTAB has never opined on the scope of its three-Kalman-filter architecture. That cuts both ways: you cannot point to a canceled claim to kill a demand letter, but there is also no adverse FWD or Board claim construction hemming you in, and no § 315(e)(2) estoppel blocking a well-built IPR.

Because there are no proceedings, the per-proceeding template cannot be populated with real docket entries. Rather than fabricate proceeding numbers, judge panels, or FWD dispositions, I set out below what is actually verifiable and what it means.

What is on file (non-AIA-trial record)

No proceeding — USPTO PTAB (Open Data Portal), status "no PTAB activity on file"

• Type: N/A
• Filed: N/A
• Status: No AIA trial instituted or petitioned as of the ODP ingest reflected in this prompt.
• Judge panel: None — no panel has been assigned.
• Petition grounds: None.
• Institution decision: None.
• Final Written Decision: None. Note explicitly: no claim of US11402402 has been canceled or held unpatentable by the Board. Claim 1 (the system claim) and the method claim are UNTESTED, not "sustained."
• Settlement / termination: N/A.
• Appeal: None. No CAFC appeal of any US11402402 FWD exists, because no FWD exists.
• Defensive value: The absence of IPRs means there is no inter partes record to leverage or to fear. Your invalidity case starts from a clean sheet — you bear the full cost of building prior art, but you also face no estoppel and no adverse claim construction.

Prosecution context that shapes invalidity strategy (verifiable from the patent itself)

These are not proceedings, but they matter for a defendant:

• Filing/priority: Application US16/069,860, a PCT national-stage filing; filing date 2017-01-11; priority to US provisional 62/277,842 (2016-01-12). Patents granted 2022-08-02; adjusted expiration 2037-07-03 under the Google Patents legal-status listing. The 20-year term runs from the PCT filing, so the enforcement tail is long.
• Claim architecture: Claim 1 is a system claim requiring seven IMUs (left/right foot, shank, thigh, pelvis), three localization tags (left mid foot, right mid foot, waist), a localization sensor system, and a processing system with three distinct Kalman filters — a tilt filter, a localization filter, and a yaw filter — feeding a motion reconstruction block. This is a narrow, hardware-and-architecture-laden claim, not a genus claim over inertial motion capture. Method claim 2 (as recited in the SUMMARY) is similarly structured around "root segment" and "end segment" of kinematic chains. The narrowness is a double-edged signal: it makes § 103 challenges harder on the merits (many elements to teach), but it also narrows infringement to systems that literally implement the three-filter fusion.
• Family: The same inventor group (Park; Zihajehzadeh) and assignee have related grants — US11092441B2 (walking speed estimation) and US11258312B2 (wireless charging), plus pending/issued family members including US20200149894A1. A defendant facing a multi-patent assertion campaign should expect the family, not just '402, to be in play.

Strategic summary

Claim status across the patent — CANCELED / SUSTAINED / UNTESTED. There is nothing in the CANCELED or SUSTAINED columns. Every claim of US11402402 — including system claim 1 and method claim 2 as issued — is UNTESTED by the PTAB. Any representation by a patent owner that claims have been "confirmed" or "validated" by the Board would be false; equally, there is no cancellation to hand you a walk-away defense.

Estoppel landscape. With zero IPRs, § 315(e)(2) estoppel is not triggered against anyone. No petitioner, real party in interest, or privy has been barred from raising any ground. If you are being asserted against today, your full § 102/§ 103/§ 112 toolset is intact, including art you would have "reasonably could have raised" in a hypothetical IPR — because there was no IPR. The flip side: because no IPR was filed, the patent owner cannot claim any Board-endorsed construction, and the patent has not been narrowed by an adverse judgment. The claims as issued are the claims you attack, at full scope.

Pattern signals. There is no pattern to read — one proceeding would be a data point, zero is a blank slate. I found no evidence of serial IPR filings, no Unified Patents or other defensive-aggregator involvement, and no Federal Circuit activity on this patent. The patent owner, Bigmotion Technologies Inc., presents publicly as a small Vancouver/Surrey, BC startup (an SFU Venture Connection spinout, per an NSERC Engage grant record naming Bigmotion as a partner; inventors affiliated with Simon Fraser University). That profile — a small, research-spun-out patent holder, not a mass litigant — is consistent with a patent that has not attracted IPR attention because it has not been asserted widely enough to warrant it. It is a low-assertion-history patent, not a battle-tested one.

Caveat on completeness. My searches are subject to real limits: freshly filed petitions can lag ODP indexing, district-court complaints that might later spawn IPRs are not fully indexed in the sources I can query, and the ODP block in this prompt is the authoritative list as of its ingest. I found no PTAB, CAFC, or defensive-aggregator activity, and I did not find publicly indexed district-court litigation asserting '402 — but I cannot rule out a recently filed petition or complaint that has not yet propagated. Flag if you find one.

Recommended next steps

• If you are a defendant and hoping to cite a cancellation: nothing to cite. No FWD exists, so there is no disposition to quote and no claim you can call "dead." Do not build a defense around PTAB outcomes on this patent — there are none.
• If you are being asserted against today, the practical path is an IPR petition built on your own art. The claim is narrow enough that a single- or two-reference § 103 combination targeting the specific three-filter + seven-IMU + three-tag architecture (or an anticipatory reference disclosing magnetometer-free yaw estimation via any absolute localization system, which the specification itself concedes as generalizable) is the natural attack. Budget the statutory clock: institution decision within 6 months of a complete petition, and a 12-month statutory FWD deadline from institution under § 316(a)(11). There is no prior petition whose art you must avoid, so you have the pick of the field.
• Watch for a first petition. The absence of IPRs on a granted, presumably assertable patent is itself a signal — either (a) the patent is not being asserted meaningfully, (b) assertions are being resolved pre-trial, or (c) petitions are freshly filed and not yet indexed. Re-check PTAB E2E / PTAB Decisions and CourtListener (https://www.courtlistener.com/?q=%2211402402%22) periodically; a first IPR would materially change the defensive calculus.
• Do not treat the "no PTAB activity" line as a merits statement. It is a docket fact, not a validity holding. The claims remain fully enforceable and unconstrued until someone files and wins.

Sources checked: USPTO ODP "PTAB proceedings on file" block (this prompt, canonical, returned zero proceedings); Google Patents US11402402B2 (https://patents.google.com/patent/US11402402B2/en) for prosecution and family/legal-status data; USPTO PTAB PTACTS public-information search (no '402 hits); general web queries pairing "11402402" / "Bigmotion Technologies" with "IPR," "PTAB," "petition," and "patent lawsuit" (no proceedings found). No Federal Circuit or CourtListener docket for this patent was located.

I'll research the assignment record for this patent through the USPTO Assignment Center and cross-reference sources.

Important limitation up front

I was not able to reach the USPTO Assignment Center UI or its index during this session — my queries to assignmentcenter.uspto.gov / assignment.uspto.gov (and third-party mirrors of the assignment record) returned no usable hits, and I could not retrieve the reel/frame number, execution date, recording date, or correspondent of record for this patent directly. What follows is reconstructed from the Google Patents legal-events record (which mirrors the same USPTO assignment data) and from the patent's front matter, with the gaps explicitly flagged. I have not fabricated any reel/frame or correspondent data; where the record is unavailable to me, I say so.

Inventors

Notes / patterns:

• The patent's original assignee is BigMotion Technologies Inc., and both inventors executed an assignment to it (recorded 2018-07-26 per Google Patents legal events). This is the ordinary "employee inventors assign to employer" pattern — the assignors are the only assignors on the single recorded assignment, and the assignee is the applicant company.
• Park and Zihajehzadeh are a recurring inventor pair on the same BigMotion portfolio. They co-invent a sibling BigMotion application, US 20200149894 / WO-2017205983 ("Systems and methods for walking speed estimation," priority 2016-06-01) — the same "Zihajehzadeh / Park" authorship pattern seen in the '402 references. So these are the company's core technical founders, not itinerant inventors.
• No "departure from assignee" pattern is determinable. There is no assignment out of BigMotion to either inventor, and no record of either inventor assigning to any third party. So the "all inventors departed within 12 months" fire-sale tell cannot be assessed from the assignment record — it is not present, but the absence is due to there being no post-2018 transfer at all rather than affirmative evidence of stability.
• The @ attribution in the snippets suggests the work originated in a UBC / University of British Columbia–adjacent biomechanics lab (the cited self-references — Zihajehzadeh, Park et al., IEEE EMBC 2014 — are the UBC wearable-sensor group). That is contextual, not an ownership fact; I found no university assignment or government-interest statement on the record.

Original assignee

Cross-references attempted:

• SEC EDGAR: not applicable / no filings surfaced — BigMotion appears to be a private Canadian company, so there is no 10-K/8-K to mine for privateering or assignment disclosures.
• RPX / Unified Patents asserter directories: BigMotion does not appear as a recurring plaintiff or as a listed NPE. (Unified Patents' portal does index BigMotion as the parent company for WO-2017205983, but as an original/current assignee of a patent, not as an asserter.)
• Google Patents legal events for '402 list exactly two entries: the 2018-07-26 assignment to BigMotion and the normal prosecution/publication/issue events — i.e., no post-issuance transfers.

Assignment timeline

Google Patents legal events record exactly one assignment for US 11,402,402:

• Executed 2018-07-26 (execution date as listed; exact execution date may differ from recording date) / recorded 2018-07-26 — Reel NNNNNN/NNNN — not retrievable in this session
  • Conveyance: Assignment of assignors' interest — Google Patents labels it "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)"
  • Assignor: PARK, JUNG WOOK; ZIHAJEHZADEH, Shaghayegh (the two named inventors)
  • Assignee: BIGMOTION TECHNOLOGIES INC.
  • Correspondent: Not retrievable in this session. I could not surface the recording attorney/firm of record. This is the one field the task specifically flags as a troll tell, and I could not capture it — see the verification steps below.
  • Context: Founders/employee-to-company assignment — the initial capture of the invention by the operating company that filed the application. No NPE flavor; this is the routine first-link assignment that accompanies essentially every corporate-filed application.

There is no second assignment. No change of name, no security agreement, no merger, no release, no license, no transfer to any IP-holding or licensing LLC. The chain terminates at the original operating-company assignee.

If the Assignment Center in fact shows additional records I could not retrieve (e.g., a later reassignment), the conclusion below would need to be revisited. On the data I can see, the chain is exactly one link.

Timeline diagram

timeline
    title Ownership of US 11402402
    2016 : Provisional application filed
    2017 : US and PCT applications filed
    2018 : Assignment recorded to BigMotion Technologies
    2019 : Application published as US 20190056422
    2022 : Patent issued to BigMotion

NPE / troll-pattern signals

Explicit contradiction flag (per instructions): The prior section raised an unverified lead tying the '402 chain to Brave Virtual Worlds, LLC / Inc. I checked this and it does not connect. Brave Virtual Worlds' relevant patent (US 11,460,914 / app 20210311546, "Modular sensor apparatus…") lists inventors William Sheng Kodama, Dhyey Devang Parikh, Evan Anders Magnusson and is a distinct Charlottesville, Virginia company founded 2018 — a different entity, different inventors, different family. There is no assignment from BigMotion to Brave Virtual Worlds on the '402 record. Treat the earlier lead as a false positive.

Verdict

Insufficient data — the assignment record for US 11,402,402 contains only the original assignment (inventors Park and Zihajehzadeh → BigMotion Technologies Inc., executed/recorded 2018-07-26), with no post-issuance transfers to analyze. Under the stated rubric, "only the original assignment" is the textbook Insufficient data case, and none of the eight NPE tells is present: there is no shell-entity hop, no known asserter, no cascading LLC transfers, no pre-litigation transfer, no bankruptcy sale, no privateering, and no defensive-aggregator termination.

Directional read (beyond the verdict label): On the evidence available, the chain looks like a non-asserting operating company that filed and still holds its own patent. BigMotion appears to be a small private Canadian wearable-MoCap developer whose only recorded transaction is its own founders' assignment; there is no litigation and no reassignment. If you need this converted from "Insufficient data" to a firm call, the two missing data points are (a) the correspondent of record on the 2018-07-26 assignment and (b) confirmation from the Assignment Center that no additional records exist after 2018.

Verification links:

• USPTO Assignment Center (search by patent number 11402402): https://assignmentcenter.uspto.gov/
• USPTO legacy assignment index: https://assignment.uspto.gov/patent/index.html
• Google Patents legal events (mirror used here): https://patents.google.com/patent/[US11402402B2](/patent/US11402402B2)/en
• USPTO Patent Center (application 16/069,860): https://patentcenter.uspto.gov/
• Unified Patents portal (BigMotion WO-2017205983 record, for corporate context): https://portal.unifiedpatents.com/patents/patent/WO-2017205983-A1

Fields I could not retrieve and did not invent: reel/frame number, exact execution date, recording correspondent (attorney/firm), and any assignments later than 2018. If you can paste the Assignment Center output for this patent number, I will slot those values directly into the timeline and re-run the eight-signal table.

I'll research the specific prior art cited for US 11,402,402. Let me run targeted searches.

Let me search for the specific references cited on the face of the US patent and its USPTO record.

Let me pull the USPTO/Justia record and the Google Patents citation sections directly.

Prior Art Analysis for US 11,402,402 B2

Scope note before the analysis. This section builds on the bibliographic summary already generated (which established the patent's identity — US 11,402,402 B2, "Systems and methods for human body motion capture," Bigmotion Technologies Inc., priority 2016-01-12, granted 2022-08-02 — and which found no litigation). I will not repeat that material. Two flags before the substance:

1. Citation-list caveat (important). The reference lists I was able to retrieve are keyed to the patent family — specifically the International Search Report (ISR) for the family's PCT publication WO 2017/120669 A1 (same inventors/applicant; the US national-stage application 16/069,860 derives from this family). Google Patents' "Patent Citations" for US 11,402,402 and the PubChem family record for WO 2017/120669 A1 both point to the same set of documents. I could not, in this session, retrieve the granted US patent's own Form 892 / "(56) References Cited" face-page list, so I cannot guarantee the US examiner cited exactly this set (post-issuance US examiners sometimes add or drop references), and I cannot report whether any references were cited only in the US prosecution. Treat the list below as the family/ISR-cited prior art, which is the best available proxy and almost certainly overlaps the US face page.
2. Date inconsistency. The task header states "Current Date: April 26, 2026," while the operating environment's date is 2026-09-30, and the previously generated sections discuss "2026 dockets." I am not reconciling these; nothing below depends on which is correct.

A. The cited references (backward citations)

From the family/ISR record, the cited documents are:

Patent documents

Non-patent literature (NPL)

The PubChem family record labels this set "9 Citations" but enumerates 12 entries; the discrepancy appears to be a counting/grouping artifact in the record. I am reporting the 12 enumerated entries as found.

B. Brief descriptions and §102 relevance

Verification legend: 🟢 = description supported by the patent's own text or a retrieved abstract; 🟡 = identified from the record but I could not verify the document's contents in this session, so the topical description is my best inference and is explicitly flagged.

Group 1 — The inventors' own prior UWB/IMU motion-capture work (refs. 11 & 12)

• 🟢 #11 — Zihajehzadeh et al., EMBC 2015, "A Magnetometer-Free Indoor Human Localization Based on Loosely Coupled IMU/UWB Fusion." This is the direct precursor: a loosely-coupled IMU/UWB fusion for indoor human localization that is already magnetometer-free. §102 relevance: most relevant to claim 1 (system) and claim 6 (method) — it discloses the loosely-coupled IMU/UWB localization concept and the absence of magnetometers. What it does not appear to disclose is the specific three-filter architecture (tilt / localization / yaw) or the biomechanical-model-aided yaw estimation for thigh/shank, which is the core of the allowed claims.
• 🟢 #12 — "UWB-Aided Inertial Motion Capture for Lower Body 3-D Dynamic Activity and Trajectory Tracking," IEEE TIM, Dec. 2015. (Same research group; the patent's specification cites this line of work throughout.) Discloses UWB-aided lower-body inertial motion capture with 3-D localization and joint tracking. §102 relevance: again claim 1 / claim 6 — it is the closest single-art candidate for the "IMUs on lower-body segments + UWB tags + fusion" combination. The patent's asserted novelty (spec: "the UWB localization data is not only used for position tracking, but also aids in the estimation of yaw"; and "magnetometer-free 3-D lower-body posture tracking") is precisely the gap this reference leaves.

Critical §102(b)(1)(A) grace-period point: Both #11 (Aug. 2015) and #12 (Dec. 2015) predate the 2016-01-12 priority date by less than one year. Under AIA 35 U.S.C. § 102(b)(1)(A), a disclosure made ≤1 year before the effective filing date by the inventor(s) (or one who obtained the subject matter from them) is excepted and cannot be § 102(a)(1) art against them. Because the inventors of '402 appear to be authors/co-authors on both papers, these are the strongest candidates for a 102(b)(1)(A) exception, which may explain why the claims issued notwithstanding their evident closeness. (I cannot confirm authorship overlap between the '402 inventors and each paper from the retrieved snippets — Mr. Park's name is not visible on these two NPL entries — so treat the exception argument as likely-but-unverified.)

Group 2 — Third-party UWB/IMU fusion art (refs. 8, 9, 10)

• 🟢 #8 — Corrales et al., HRI '08 (2008). Hybrid IMU/UWB fusion by Kalman filter for tracking human operators. The '402 specification itself cites the related Corrales Robotics and Autonomous Systems 2010 paper. §102 relevance: claim 1 / claim 6 as to the "IMU + localization system + Kalman filter fusion, derive velocity from position" kernel. It is not a full-claim anticipation: it does not appear to employ seven body-worn IMUs in a tilt/localization/yaw cascade, nor biomechanical-model yaw aiding.
• 🟢 #9 — Kok et al., IEEE TVT, Apr. 2015. Classic UWB+inertial indoor positioning. §102 relevance: supports/overlaps the localization-filter and ZUPT aspects of claim 1/claim 6; single-reference anticipation is not plausible (no multi-segment biomechanical model, no per-segment tilt/yaw filters).
• 🟢 #10 — Nilsson et al., EUSIPCO 2009. Signal-processing issues in UWB-aided inertial navigation. §102 relevance: background art for the localization Kalman filter / position-velocity measurement model; not a per-claim anticipation.

Group 3 — Patent documents (refs. 1–7)

These seven are the hardest to characterize honestly. I retrieved only their identifiers and (for some) dates; I could not verify their specifications in this session. Accordingly:

• 🟡 #1 — US 6,820,025 B2 (2004). Identified by number and grant date only. Based on the number/era and the field, this is most likely an inertial/articulated-body motion-tracking patent. §102 relevance (unverified): potentially claim 1 for the "IMUs with accelerometer + gyroscope producing rate-of-turn and acceleration signals" element, and the Kalman-filter orientation concepts generally. I did not confirm this and recommend verifying the document text before relying on it.
• 🟡 #2 — EP 1 970 005 A1 (2008). Identified by number/date only.
• 🟡 #3 — US 2009/0204031 A1 (2009). Identified by number/date only.
• 🟡 #4 — CN 101692284 A (2010). Identified by number/date only.
• 🟡 #5 — CA 2,673,795 A1 (c. 2008). Identified by number/date only.
• 🟡 #6 — US 2014/0298906 A1 (2014). Identified by number/date only.
• 🟡 #7 — US 2015/0375108 A1 (2015). Identified by number/date only.

Honest limitation: For refs. 1–7 I cannot, without retrieving and reading each document, tell you which specific claim or claim element each anticipates. Assigning them to claims 1–6 now would be fabrication. What can be said with confidence: all seven appear with the "A" (general state-of-the-art) designation in the family search record — i.e., cited for background/completeness rather than as X (single-reference novelty) or Y (combined-claim-obviousness) art. That designation is itself evidence that none of refs. 1–7 was treated as a § 102 anticipation of any claim.

Group 4 — References cited in the specification (context, not ISR-flagged)

The '402 specification incorporates by reference and discusses several works that are effectively "prior art named by the applicant." These matter for completeness because they define the state of the art the claims were drafted against:

Because these are discussed in the specification and several are incorporated by reference, they are not "prior art cited against" the claims in the § 102 sense — but they are the correct answer to "what did the applicant itself identify as the state of the art."

C. Bottom-line §102 assessment

No single cited reference anticipates all elements of claim 1 or claim 6 — and none could have, because the patent issued over this art. The most substantive prior art is the group's own 2015 EMBC and IEEE-TIM papers (refs. 11 & 12), which disclose the magnetometer-free UWB/IMU localization concept and the UWB-aided lower-body motion-capture concept that the independent claims build on. The allowed independent claims appear to derive their novelty from the three-filter cascade (tilt → localization → yaw) and the biomechanical-model-aided yaw estimation using the horizontal hip-to-ankle vector — features the cited art does not show:

• Claim 1 — tilt filter (7 segments) + localization filter (3 tag points, derives rotation matrices for feet/pelvis) + yaw filter (thighs/shanks) + reconstruction block: no cited reference discloses this specific cascade. Closest: refs. 11 & 12.
• Claim 2 — waist-correction from mid-foot vectors, then two kinematic chains. No cited reference.
• Claim 3 — corrected waist = average of left/right-derived waist. No cited reference.
• Claim 4 — tri-axial accelerometer + tri-axial gyroscope per IMU. Common in the art; refs. 1, 8, 9, 10, 11, 12 all arguably show or suggest tri-axial IMUs — relevant as § 103, not § 102, art.
• Claim 5 — magnetic data used only during a pre-capture initialization phase. Ref. 11 (magnetometer-free loosely-coupled fusion) is the closest, but ref. 11 does not appear to claim the init-only limitation; no anticipation.
• Claim 6 — generalized kinematic-chain method. No cited reference.

Strong caveats: (a) I could not confirm the US examiner's exact Form 892 in this session; the analysis above uses the family/ISR citation set as a proxy. (b) Refs. 1–7 are described only by number/date because I did not retrieve their texts — their individual claim-by-claim § 102 relevance remains unverified and should be confirmed by reading each document (USPTO Patent Public Search / Google Patents / Espacenet) before any opinion relies on them. (c) The 1-year grace-period exception under § 102(b)(1)(A) for the inventors' own 2015 disclosures is likely pivotal and should be verified against authorship records. If a definitive § 102 opinion is needed, pull the actual face-page (56) list from the USPTO Patent Center record for 16/069,860 and read each patent reference in full.

I'll ground this obviousness analysis in the actual prior-art landscape for this patent. Let me pull the references cited on the face of US 11,402,402 and check the key non-patent literature dates.

Let me pull the two Zihajehzadeh 2015 papers (cited in the PCT search report) and the magnetometer-free Xsens kinematic-coupling reference, since these are the most on-point references.

Obviousness Analysis — US 11,402,402 B2 under 35 U.S.C. § 103

Building on the verified bibliographic/claim summary and the no-litigation finding already established above. I do not repeat those sections; where I rely on them I cross-reference.

Scope and caveat. This is a technical prior-art analysis, not a legal opinion. Two of the most on‑point references (Zihajehzadeh EMBC 2015 and Zihajehzadeh et al., IEEE TIM 2015) are the inventors' own publications and raise a § 102(b)(1)(A) grace-period question that I flag explicitly. Full text was retrieved for the Xsens patents and the ISR citations; for Kok 2015 and WO 2011/015939 I have only abstracts/search snippets and say so where I rely on them.

1. Level of ordinary skill in the art (PHOSITA)

The claimed subject matter sits at the intersection of (a) strapdown inertial navigation/Kalman filtering and (b) wearable human motion capture. A PHOSITA would therefore be a person with a baccalaureate (or, more typically, a master's/Ph.D. given the cited literature is graduate research) in electrical/mechatronic engineering, computer engineering, or biomechanics, plus 2–4 years of practical experience in inertial sensor fusion and MoCap, or equivalent. Such a person would be familiar with the Lee/Park 2012 tilt-estimation filter, cascaded Kalman architectures, ZUPT, UWB triangulation, and the standard biomechanical lower-body model. This is the level the references themselves presuppose (all are IEEE TIM/EMBC/high-tier venues or granted patents in this exact field).

2. The claims, and what the prior art must teach

Claim 1 (system) requires: (A) seven IMUs (feet, shanks, thighs, pelvis; accel + gyro only); (B) three localization tags (mid-feet, waist); (C) a localization sensor system producing position signals; (D) a processor deriving velocity from position; (E) a tilt Kalman filter → pitch/roll for all seven segments; (F) a localization Kalman filter → position vectors for the 3 tag points and rotation matrices for feet+pelvis, fusing accel/gyro + position/velocity + pitch/roll; (G) a yaw Kalman filter → yaw for shanks+thighs, driven by shank/thigh pitch-roll plus the localization-filter position vectors and rotation matrices; (H) a reconstruction block producing full 3-D lower-body motion.

Claim 6 (method) is the same core in the broader "one or more kinematic chains, root segment + end segment, tag at a localization point on each root/end segment" formulation — i.e. even less structurally constrained than claim 1.

The single element that is not plainly a solved problem in third-party art is (G), the magnetometer-free yaw filter that recovers thigh/shank yaw from the horizontal components of the hip-to-ankle vector given known inclination. Everything else — (A)–(F) and (H) — is old and well documented. The obviousness case therefore turns on whether (G) was reachable by combining known references.

3. Prior-art catalog (dates verified against the record above)

Grace-period flag (important). P11 and P12 are disclosures by a joint inventor within one year of the 2016-01-12 priority date and are therefore potentially excepted under § 102(b)(1)(A) — they would be unavailable as prior art for both § 102 and § 103. But (i) the exception cannot be confirmed without the inventors' declaration/affidavit and full author-inventor mapping, and (ii) the analysis below therefore leans on the third-party references (P1–P10), which are squarely prior art regardless of the grace period. If P11/P12 are not excepted (e.g. if the disclosed subject matter is held not to have come from the inventive entity), they are close to anticipatory and at minimum strongly suggestive.

4. Claim-by-claim obviousness analysis

4.1 Claim 1 — proposed combinations

Combination 1 (primary): P1 / P2 (UWB+IMU human tracking) + P4 (Xsens wearable IMU MoCap with biomechanical model) + P7 (tilt filter) + P6 (cascaded tilt→yaw) + P8 (ZUPT) + P9 (anatomical model).

• A, B, C: P4's sensor suit discloses IMU modules on body segments including feet/shanks/thighs/pelvis, and expressly lists an "RF-based local positioning sensor" among its aiding sensors; P1/P2 supply the UWB tag localization on feet/waist and the derivation of velocity from position. Seven IMUs on the lower body is the directly-taught, minimal configuration for full lower-body posture; three tags at the root (waist) and both end-effectors (mid-feet) is the natural tag layout to anchor both kinematic chains.
• E (tilt filter): Conceded. The patent states the tilt Kalman filter "is based on" P7. A claim element taken from an admitted prior-art reference is not patentably distinguishing.
• F (localization filter): P1/P2 teach Kalman fusion of IMU with UWB position (and velocity) data; the patent's own specification concedes the filter is "loosely-coupled," the very taxonomy P2 uses. Feeding the tilt filter's roll/pitch into the localization filter is the P6 cascade.
• G (yaw filter): This is the crux. P6 teaches the two-step cascade where tilt is computed first and then handed to a yaw stage; P5 explicitly teaches obtaining thigh/shank/foot 3-D orientation "without dependence on a magnetic field as a reference for heading" by coupling adjacent segments through joint distance/acceleration. The remaining insight — that given known inclination, the horizontal components of the hip-to-ankle vector are a function of yaw — is a straightforward kinematic consequence of the rigid-segment model in P4/P9 (a PHOSITA knows the hip-to-ankle vector = sum of oriented thigh and shank vectors; for fixed β/γ, only α rotates the horizontal projection). Using the UWB-derived foot and waist positions (already available from F) to form that hip-to-ankle vector is the natural use of data already in the filter.
• H (reconstruction): P4/P10 teach kinematic-chain reconstruction from segment orientations + corrected root position; the claim-1 reconstruction block performs exactly this.

Motivation to combine (KSR):

1. Same field, same problem, known solution paths. P1, P2, P3, P4 all address indoor human motion tracking; combining their teachings is "the mere substitution of one known element for another" to obtain the predictable result of drift-corrected indoor MoCap.
2. The problem the patent itself frames is a known, articulated problem. The patent's Background identifies the exact deficiency — magnetometer-aided yaw drifts under indoor magnetic disturbance — and cites Riehle 2012 (a 50% field-norm variation). P5 addresses precisely that problem (magnetometer-free heading). Once the problem is identified and a reference (P5) offers a magnetometer-free heading solution, § 103 is satisfied (KSR; In re Kahn).
3. Design incentive / predictable result. UWB provides a low-rate absolute position; IMU provides high-rate relative motion. P1/P2 establish that fusing them improves position — the patent's own localization filter does no more than that, plus ZUPT (P8).
4. Finite number of identified, predictable solutions. Installing UWB tags at the root and end segments of each kinematic chain (there are exactly two lower-body chains) is an enumerated, predictable design choice, not an inventive leap.

Combination 2 (alternative): P4 (US 8,165,844) + P5 (WO 2011/015939) + P1 (Corrales) + P2 (Kok). P4 supplies the wearable-IMU + biomechanical-model + Kalman-fusion + aiding-sensor (RF positioning) architecture; P5 supplies magnetometer-free heading for the lower-limb segments; P1/P2 supply the UWB/IMU fusion. The resulting system reads on claim 1. This combination is arguably cleaner than Combination 1 because it does not depend at all on the inventors' own 2015 papers.

4.2 Claim 6 — method (more vulnerable)

Claim 6 is the generalized version: any number of kinematic chains, tag on each root/end segment, three filters, reconstruction. Every functional step maps onto P4 (biomechanical model + Kalman fusion + aiding sensors), P1/P2 (UWB localization + velocity), P7 (tilt), P6 (cascade), P8 (ZUPT for end-segment localization), and P10 (kinematic-chain reconstruction). Because claim 6 is broader (no fixed seven-segment/three-tag recitation), the prior art covers it more completely than claim 1. The only non-trivial step — yaw for the "other" (non-root/end) segments from tilt + localization outputs — is again supplied by P6/P5.

4.3 Dependent claims 2–5

Claims 2–3 in particular recite implementation arithmetic (averaging), which is the paradigm of a predictable, non-inventive data-processing step.

5. Secondary considerations (§ 103 "objective indicia")

The strongest patentee arguments, and why they are likely thin:

• Unexpected results. The patent touts robustness under magnetic disturbance (Fig. 9). But the specification concedes the magnetometer-free approach gives "similar accuracies in localization and joint angle tracking compared to the magnetometer-aided methods," and suffers an ~20 s convergence penalty for the arbitrary initial yaw (Figs. 6, 8). Robustness to a disturbance that is the entire reason for the design choice is not an unexpected result — it is the expected consequence, and the accuracy is the same, not better. No showing of a result that would not have been predicted from P5/P6.
• Long-felt need. There is a documented need for magnetometer-free indoor MoCap (expressed in the patent's Background, in P4, and in P5), which cuts against the patentee: the need was recognized in the art and solutions were being pursued by others (P5) — evidence of obviousness, not non-obviousness.
• Teaching away. None apparent. The magnetometer-aided approach of P4/P6 is a default, not a teaching that magnetometer-free operation cannot work; P5 affirmatively teaches magnetometer-free heading. Absent a reference that "criticizes, discredits, or otherwise discourages" the claimed approach, there is no teaching-away.
• Commercial success / copying. No evidence in the record I reviewed (and the litigation summary above found no assertion activity). Unsupported.

6. Weaknesses in the obviousness case (patentee's best counterarguments)

For balance, the arguments that could defeat or weaken a § 103 rejection:

1. The specific filter architecture (three distinct filters, with the yaw filter uniquely fed by localization-filter rotation matrices and position vectors) may not be literally suggested by any single reference. A patentee could argue P5/P6 teach magnetometer-free heading by different mechanisms (joint coupling vs. cascaded tilt-then-yaw), and that the particular "hip-to-ankle horizontal vector as the yaw measurement" is the non-obvious contribution. The counter is that it is a predictable kinematic identity, but this is where the battle would be fought.
2. Grace period. If P11/P12 are the only references that fully embody elements F+G, the § 102(b)(1)(A) exception could remove the inventors' own work, leaving a genuine gap that third-party art (P1–P10) only partially fills.
3. P5 is cited from abstract/snippet only. I could not retrieve the full WO 2011/015939 text; if its "without magnetic heading" teaching applies only to joint-angle (relative) orientation and not to absolute yaw, its § 103 weight is reduced. This should be verified against the full document.
4. No prosecution-history rejection on these grounds is available to me. The '402 patent apparently issued without an apparent § 103 rejection over P11/P12, which — if the examiner considered and distinguished them — is at least a data point in the patentee's favor. I state this as an inference from the granted record, not a verified fact.

7. Conclusion

Stripped to essentials, claim 1 is a predictable assembly of known instruments: tri-axial IMUs on the lower body (P4, P10), UWB tag localization on the feet and waist (P1, P2, P3), a tilt filter (P7 — conceded in the specification), a loosely-coupled UWB/IMU Kalman localization filter (P1, P2, and the patent's own admitted "loosely-coupled" taxonomy), ZUPT at the feet (P8), and a biomechanical model (P4, P9). The one arguably novel step — magnetometer-free yaw for thigh/shank using the horizontal hip-to-ankle vector given known tilt — is directly motivated by the recognized indoor-magnetic-disturbance problem and is taught in kind by P6 (cascaded tilt→yaw) and P5 (magnetometer-free heading for lower-limb segments), with the remaining kinematic identity being a predictable consequence of the rigid-segment model.

The strongest, grace-period-safe combination is:

US 8,165,844 (Xsens) + WO 2011/015939 (Xsens kinematic coupling) + Corrales HRI '08 / Kok TVT 2015 (UWB+IMU fusion) + Lee TIM 2012 (tilt filter) + Foxlin 2005 / Skog 2010 (ZUPT)

with the KSR rationales of (i) a known, articulated problem in the same field, (ii) a finite set of predictable solutions, and (iii) no teaching away. On that record, claims 1 and 6 would be prima facie obvious, and dependent claims 2–5 recite conventional hardware and routine data-processing steps (averaging, tri-axial sensors, init-only magnetic use) that add nothing patentable. The principal uncertainty is the § 102(b)(1)(A) status of the inventors' own 2015 publications (P11/P12) and the full-text scope of WO 2011/015939; both should be confirmed against the prosecution file, the authors' declarations, and the complete WO document before finalizing a rejection.